SOLAR ACTIVITY — SPENCER JONES 237 



accompany a storm are of great complexity and although they can 

 be accounted for in their broad outlines, there is not as yet any com- 

 pletely satisfactory theory of magnetic storms. 



There are some significant differences between gi-eat magnetic 

 storms and smaller storms. A great storm is almost always associated 

 with a large sunspot and usually occurs within a day or two of the 

 passage of the spot across the central meridian. It is often preceded 

 about a day previously by a bright flare. 



The smaller storms may be, but usually are not, associated with a 

 sunspot. They may occur when no spot is to be seen on the disk. 

 Like the great storms, the small storms must be produced by corpus- 

 cular emission from the sun ; the emission must sometimes, therefore, 

 take place from regions where there is no visible spot. The frequency 

 of great storms correlates very closely with the sunspot frequency; 

 the smaller storms, however, reach a maximum a year or two after 

 sunspot maximum and show a pronounced lag, as compared with the 

 sunspots, during the period of decreasing sunspot frequency. Small 

 storms often occur within the two years preceding sunspot minimum, 

 when sunspots are few. 



Both magnetic storms and aurorae show a tendency to recur after 

 27 days, which is approximately the period of rotation of the sun 

 relative to the earth. This is to be expected if the particle radiation 

 from the sun is produced by a cause which persists for longer than 27 

 days. A great storm, however, is very rarely followed by appreciable 

 disturbance after a 27-day interval, while a small storm may have 

 several recurrences at 27-day intervals. The small storms are on the 

 whole longer lived than the great storms. It seems then that the 

 mechanism which gives rise to the small storms is essentially different 

 from that which gives rise to the great storms. The small storms 

 mostly come from areas of long-lived disturbance. A connection with 

 filaments which, as we have seen, are also long-lived, has been suspected. 



Another solar phenomenon which is intimately linked with large 

 sunspots, solar flares, and radio fadeouts is the emission by the sun of 

 radiation in the centimeter and meter wavelengths. During the solar 

 maximum of 1937-38 many radio amateurs reported a troublesome 

 hiss in their 10-meter receivers, which they found to be associated with 

 periods of solar activity. At various times during the last war radar 

 transmitting and receiving installations recorded strong sources of 

 noise which were at first thought to arise from enemy interference but 

 which, on further investigation, were found to occur only when the 

 aerial was directed toward the sun. As the sun radiates approxi- 

 mately as a black body with a temperature of about 6,000° K., some 

 radiation in radio wavelengths is to be expected. But the intensity 

 of such radiation would be very far below that which could be detected 

 with the most sensitive short-wave receivers available. 



